Many chemical analyses are background-limited, that is, the lower limit of their applicable range is set by uncertainties in establishing the position of the background. Minimizing this uncertainty leads to useful improvements in analytical performance. This dissertation examines and compares digital data processing techniques which can be used for this purpose. Chapter 1 briefly reviews the fundamental aspects of data processing. In the subsequent Chapters, four projects involving background correction are then discussed. / The success of background correction by data processing depends mainly on the efficiency and reliability of the programs used. Attempts to improve these two important factors have mainly dealt with programming languages and with techniques of overall program design. The concept of reusable software will be introduced and its value, particularly in a research environment, will be discussed in Chapter II. / Chapter III deals with the qualitative aspects of background correction. Its goal is to improve spectral information. Only minimal spectral distortion is allowed. Five data processing techniques are described and their performance on simulated data sets, with better control of factors such as signal-to-noise ratio (S/N) and background conformation, is examined. The relationship between the design philosophy and correction efficiency is discussed. / Chapter IV deals with the quantitative aspects of background correction. Its goal is to improve the quality of quantitative information. Fourteen data processing techniques are described and evaluated. An attempt has been made to validate the procedures used in constructing the simulated sets. They are closely modeled on the Raman spectrum of dimethyl sulfone which is a suspected intermediate of sulfuric acid rain cycle. The optimal results show a compromise between noise amplification and background suppression. / Chapter V is an extended application of the optimal methods, developed in Chapter IV. They are applied to an infrared spectrum of a mixture of three classes of serum lipid which are of clinical interest. It is shown that these techniques are still effective in severe peak overlap cases, thus expanding their application range. / Source: Dissertation Abstracts International, Volume: 44-11, Section: B, page: 3394. / Thesis (Ph.D.)--The Florida State University, 1983.
| Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_75233 |
| Contributors | LING, YONG-CHIEN., Florida State University |
| Source Sets | Florida State University |
| Detected Language | English |
| Type | Text |
| Format | 197 p. |
| Rights | On campus use only. |
| Relation | Dissertation Abstracts International |
Page generated in 0.0025 seconds